Shopify Scraper: Products, Prices & Store Leads

Scrape any public Shopify store in one actor:

• Products & prices: one flat row per product or variant (SKU, price, compare-at price, stock), with change detection between runs: new, price up, price down, availability, removed.
• Shopify store leads & contacts: business emails, phone numbers, company name, address and social links (Instagram, Facebook, TikTok, YouTube, X, Pinterest, LinkedIn), taken only from the shop's own pages.

Pay per event, no subscription: $0.50 per 1,000 product rows, $0.20 per 1,000 variant rows, $4 per 1,000 shops with an email or phone. Shops without email/phone or that cannot be read are free. No proxies, no browser. With onlyChanged you pay only for products that actually changed, which makes daily price monitoring very cheap.

Use cases

• Competitor price monitoring and sale detection (compareAtPriceMax = original price)
• Catalog and assortment tracking for brands and marketplaces
• Stock / availability alerts
• Product feeds for analytics, BI or spreadsheets
• Shopify lead generation: store emails, phones and socials for B2B outreach, supplier and partner research (outputMode: "contacts")

Input

{
  "storeUrls": ["https://www.allbirds.com", "kith.com"],
  "maxProductsPerStore": 100,
  "outputMode": "product",
  "onlyChanged": false,
  "previousDatasetId": "PREVIOUS_DATASET_ID"
}

• storeUrls (required): shop URLs or domains. Normalized to HTTPS and deduplicated.
• maxProductsPerStore: omit to read the whole catalog.
• outputMode: product (default), variant or contacts. Product and variant rows use separate comparison keys.
• onlyChanged: compare against previousDatasetId and output new, changed and removed products or variants; without it, deduplicate within this run.
• previousDatasetId: optional read-only comparison dataset; ignored in contacts mode.

Output

One row per product:

{
  "storeDomain": "allbirds.com",
  "productId": "7340901859408",
  "handle": "womens-allbirds-flip-flop-dusty-pink",
  "url": "https://allbirds.com/products/womens-allbirds-flip-flop-dusty-pink",
  "title": "Women's Allbirds Flip Flop - Dusty Pink",
  "vendor": "Allbirds",
  "productType": "Shoes",
  "priceMin": 25,
  "priceMax": 25,
  "compareAtPriceMax": 50,
  "currency": "USD",
  "available": true,
  "variantCount": 7,
  "updatedAt": "2026-10-06T00:06:16-07:00",
  "change": "new",
  "previousPriceMin": null
}

change is one of new, price_up, price_down, availability, unchanged, removed. Prices are numbers in the shop currency (not cents). currency comes from the shop's public /meta.json; if it is missing it stays null (never guessed).

With outputMode: "variant", each row contains storeDomain, productId, variantId, handle, url (ending in ?variant=<id>), title, variantTitle, sku, vendor, productType, price, compareAtPrice, currency, available, updatedAt, change, previousPrice. Variant fields come from /products.json; barcode/GTIN is not provided by this public feed and is not offered.

Removed rows retain the last known metadata, set available to false and prices to null, and report the last price in previousPriceMin or previousPrice. Older dataset rows without metadata return null for those fields. Removal is detected only after the complete catalog is read: never after an error, skipped shop, product limit or cost limit. Removed rows are always emitted with onlyChanged and charged at the selected mode's row price. Prior removal rows are not treated as active products in the selected baseline. Unchanged rows are skipped without charging.

Product and variant modes collect only product data. No customer data or reviews are collected.

Pricing

Event	Price
product (one output row)	$0.0005
variant (one output row)	$0.0002
store-contacts (shop with at least one email or phone)	$0.004
Actor start (per run, per GB of memory)	$0.00005

Unchanged products skipped by onlyChanged are not charged. Shops that cannot be read are not charged. Contact rows without email or phone are delivered free. The run stops cleanly when your maximum cost per run is reached.

Scheduled monitoring

Create an Apify Schedule for this actor, for example daily at 06:00 in your chosen timezone. Select a previous dataset and omit maxProductsPerStore so every run reads the full catalog:

{
  "storeUrls": ["https://shop.example"],
  "outputMode": "product",
  "onlyChanged": true,
  "previousDatasetId": "PREVIOUS_DATASET_ID"
}

Set previousDatasetId to the last run's dataset ID (select it with the dataset picker). The selected dataset is only read and is never modified. Missing, invalid or inaccessible datasets produce a log warning and a comparisonNotice in SUMMARY; the run continues with duplicate suppression within the current run. Without a previous dataset, every distinct current row is eligible for output. No persistent history is maintained. A dataset containing only new/changed rows, a filtered result or a cost-interrupted result is an incomplete baseline: rows absent from it can be emitted again. For reliable repeated monitoring, compare with a complete snapshot dataset and update the input through your scheduler or integration; a static schedule does not automatically select the last run.

Products are matched by store domain and product ID; variants additionally use the variant ID. Prices and availability are compared as before. Removal rows require a complete catalog read and an item present in the selected baseline. Previous removal rows are excluded from the baseline.

Use with AI agents / MCP

The actor works through the Apify API and the Apify MCP server, so AI agents (Claude, ChatGPT, Cursor and others) can call it as a tool.

MCP server URL for your MCP client:

https://mcp.apify.com?tools=pfistyle.digital/shopify-products

Example agent prompt: "Get the contacts of colourpop.com and tentree.com with pfistyle.digital/shopify-products (outputMode contacts)."

Direct API call that runs the actor and returns the rows:

curl -X POST "https://api.apify.com/v2/acts/pfistyle.digital~shopify-products/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" -H "Content-Type: application/json" \
  -d '{"storeUrls": ["allbirds.com"], "maxProductsPerStore": 20, "outputMode": "product"}'

How it works

The actor reads the public Shopify endpoint /products.json (250 products per page), at most one request per second per shop, with retries and backoff on 429/5xx errors.

Known limitations

Small and mid-size Shopify shops usually work

Classic Shopify stores serve the standard public /products.json feed, and most small and mid-size Shopify shops do. These work in all modes. Contacts mode does not need the product feed, so it also works on many shops where products cannot be read. Check the SUMMARY record after each run to see which shops delivered data.

Large brands: tested 2026-10-07

Large brands more often use headless storefronts (own frontend, Shopify only in the background), which do not expose the product feed.

Mode	Tested	Delivered data	No data, not charged
Products / variants	20 well-known brands	13 (65 %)	7: headless storefront, bot protection or not Shopify
Contacts	15 well-known brands	8 with email or phone (charged), 6 with social links only (free)	1 blocked (bombas.com)

Worked for products: allbirds.com, kith.com, skims.com, decathlon.com, colourpop.com, netflix.shop. Contacts also worked on headless gymshark.com.

Did not work for products:

Shop type	Example	What happens
Headless storefront	gymshark.com, fashionnova.com, ruggable.com, chubbiesshorts.com	HTTP 403 or 404 on /products.json
Strong bot protection / rate limiting	bombas.com	HTTP 429 even after retries
Not a Shopify store	heinz.com, redbullshop.com	HTTP 404
Password-protected or closed shops	–	no public feed

You are never charged for shops that cannot be read. Such shops are skipped and the run continues with the next shop. Every run writes a per-shop summary to the SUMMARY record of the run's key-value store, and the run status line lists skipped shops with their reason:

[
  {"storeDomain": "allbirds.com", "status": "ok", "products": 100, "removed": 0, "reason": null, "detail": null},
  {"storeDomain": "www.gymshark.com", "status": "skipped", "products": 0, "removed": 0, "reason": "headless", "detail": "Public Shopify endpoint unavailable (HTTP 403)"}
]

status: ok, skipped, partial (failed mid-catalog), stopped (cost or product limit reached), not_processed. reason for skipped shops: headless, not_shopify, blocked, password_protected, unreachable. products counts emitted current rows in either mode; removed counts emitted removals separately. You pay only for rows that are actually delivered to the dataset; if a shop fails in the middle of its catalog, only the rows delivered before the failure are charged.

Other limitations

• Price changes take priority over a simultaneous availability change.
• The previous dataset is an explicit read-only baseline; update its ID when advancing your monitor.

Shopify store leads: emails, phones and social links

Set outputMode: "contacts" for one row per readable shop, with no products: storeDomain, shopName, companyName, emails[], phones[], address{street, postalCode, city, region, country}, socials{facebook, instagram, tiktok, youtube, x, pinterest, linkedin}, sourceUrls[], and contactsFound.

Example row (real run, 2026-10-08):

{
  "storeDomain": "colourpop.com",
  "shopName": "ColourPop",
  "companyName": "ColourPop Cosmetics, LLC",
  "emails": ["colourpop@colourpop.com", "legal@colourpop.com"],
  "phones": [],
  "address": {"street": null, "postalCode": null, "city": "Oxnard", "region": "California", "country": "US"},
  "socials": {"facebook": null, "instagram": "https://instagram.com/colourpopcosmetics", "tiktok": "https://www.tiktok.com/@colourpopco", "youtube": null, "x": "https://x.com/colourpopco", "pinterest": null, "linkedin": null},
  "sourceUrls": ["https://colourpop.com/meta.json", "https://colourpop.com/", "https://colourpop.com/pages/contact-us", "https://colourpop.com/policies/terms-of-service"],
  "contactsFound": true
}

Sources are exclusively the shop's own /meta.json, homepage, contact-information and legal-notice policies, /pages/contact (or /pages/contact-us if unavailable), and terms-of-service only if neither a company nor email was found. At most seven requests per shop, one per second, without retries. Redirects are followed only within the shop's own domain. Missing pages are normal. sourceUrls lists successfully read sources. Unreachable shops are skipped with the reason in SUMMARY, without charges or extra diagnostic requests.

Only business contacts are extracted. Owner and director names are excluded; ambiguous company names without a legal form are omitted. Emails are normalized, deduplicated and ordered with role addresses first; system/example addresses are filtered. Socials contain profile links rather than sharing links. Address data comes from contact/legal policies, with city/region/country fallback from metadata. Extraction is conservative and may omit unstructured information.

Charged only when an email or phone is found: $0.004 per shop (store-contacts). Rows with only social links, a company name or an address are still delivered, free of charge. contactsFound is true exactly for charged rows. The event cost limit stops the run. onlyChanged, maxProductsPerStore and previousDatasetId do not affect contacts mode, and product comparison datasets are not read.

GDPR: you are the data controller and responsible for a lawful purpose and lawful use of the collected business contacts. Sources are public shop pages, never third-party databases. Personal names are not output. Keep contact datasets private.